29-30 October 2015 - Structural Reforms. Moving the debate forward

Report on DNB-IMF conference on structural reforms, held on 29 and 30 October 2015 in Amsterdam

DNB-IMF Conference

On Thursday 29 and Friday 30 October 2015, DNB and the IMF organised a conference on structural reforms in the euro area. It will be clear by now that the euro area can greatly benefit from structural reforms. The aim of the conference was to move the debate beyond this observation and assess in more detail what can be reasonably expected from reforms and what measures are most needed.
To place the topic in a context‎, in his opening words DNB President Klaas Knot highlighted the Dutch experiences with reforms in recent decades. He started with the economic downturn in the 1970s and 1980s, when the Netherlands had unemployment rates of up to14% (against the 8% average for the OECD countries) and was renowned for its employment failure. Following the Wassenaar Agreement and a series of social security reforms - in combination with a strong increase in the participation of women in the labour force  - the Dutch model in the second half of the 1990s conversely was heralded as the Dutch miracle. Incidentally, Germany showed a similar pattern - being referred to as Europe's 'sick man' in the 1990s but currently as the economic superstar. And, Klaas Knot concluded hopefully, with the economic downturn in some of the Southern European Member States – worse than in the Netherlands in the 1980s and Germany in the 1990s – conditional on the right policies we may perhaps see evidence of new employment miracles in the future.
The first session was about the macroeconomic gains we may reasonably expect from reforms. Overall, the theoretical and cross-country empirical literature agrees that the benefits of reforms in the long run are large - especially when reforms can be combined – whereas they may involve costs in the short term. Controversial issues in particular are the impact of reforms in a downturn and when interest rates are at the zero lower bound (ZLB).

Romain Duval (IMF) commenced with a presentation of a rich DSGE model. A remarkable outcome of his model is that reforms do not have deflationary results, nor does the ZLB hamper their impact. In addition, a credible announcement of reforms may have a positive effect in the case of labour market reforms, as lower dismissal costs may immediately support job creation, but not on product market reforms. To be precise, when entry costs are expected to be reduced with a delay, start-ups may postpone accessing the market. Discussant Werner Roeger (EC) pointed out several differences with the European Commission's model.  He also highlighted that the rigidities in the model could only explain 30-40% of the differences in income levels between the United States and the EU. In other words, either the list of structural rigidities is incomplete or there is more to it than just structural reforms.

DNB-IMF Conference

The second paper was by Tom Krebs (University of Mannheim). Earlier work by Krebs dealt with the impact of the recent Hartz reforms. At the request of the German Ministry of Economic Affairs, he has now been calculating the effects of six scenarios to stimulate future growth in Germany. In these scenarios the emphasis is on policies to boost well-paid, high-quality employment (instead of employment per se, as in the past). His model therefore also includes endogenous investments in human capital. His attention mainly focused on measures to smooth the marginal tax burden on labour, which currently rises sharply for jobs above the mini-job level (EUR 500 a month). According to the model, if the marginal tax burden were to increase gradually, many employees would choose to work more hours, invest more in their human capital and thus become more productive. Dicussant Andrea Colciago (DNB) drew attention to the absence of transition costs in Krebs' model, which relates to the fact that it concerns a change of the marginal tax rate and not a pure structual labour reform (in contrast, lower dismissal costs would definitely translate into short-run costs).

Finally, the paper by Christian Ebeke (IMF) was econometric. The most important motivation was to properly control for endogeneity of reforms, as they are often taken in the midst of recession and hence their impact needs to be disentangled from a 'normal' cyclical upswing. His paper first assesses the probability of a large reform (a function of economic and political variables) and then uses a local projections and matching methodology to assess the effect of a reform on the employment rate. The - very intuitive - results are that reforms have a lagged positive impact and that they are more effective when supported by fiscal and monetary policies. A less intuitive outcome of his analysis is that in a downturn, labour market reforms have a strong positive impact on employment. Discussant Máté Tóth (ECB) gave a useful summary of the method applied and placed some comments on the methodological choices, for instance the operationalisation of labour market reforms with OECD's dismissal protection index for temporary contracts.
The second session focused on a topical and controversial field of reforms, namely those concerning collective wage bargaining.

Juan Jimeno (Banco de España) presented a theoretical paper published in 2013, complemented with an empirical application to Spain. In essence, his model shows that wage negotiations at the sector level may have negative effects on employment (cf. the thesis of Calmfors & Driffil, 1988) but this negative impact disappears if there is an efficient opt-out mechanism. Consequently, the discussion should not so much focus on the optimum level of negotiations, but on the margins available to companies for adjusting to shocks. The recent Spanish reforms in the area of collective bargaining – lending priority to collective labour agreements at the company level and making opt-outs easier – should in theory provide firms with this flexibility, though Spanish companies in practice do not (yet) use these routes. According to Jimeno, this is probably because a firm-level collective labour agreement and opting-out is costly for small and medium-sized enterprises. In practice, in particular large companies tend to opt out, which is also true for the Netherlands. Another reason for firms not to use this option is that there is legal uncertainty about the measures, which means that companies might be facing high costs after all. In his discussion, Alexander Hijzen (OECD/IMF) warned about a possible downward spiral that might arise if opt-out is made easier. If companies are unhappy with the collective labour agreement and leave, which companies will be prepared to stay and stick to the collective labour agreement?

Pedro Portugal (Banco de Portugal) talked about the impact of collective bargaining in Portugal. The presentation started with a number of consecutive figures aptly showing how downward wage rigidities have disrupted the wage growth distribution in the past low-inflation years; as wages cannot go down, in 2013 three quarters of all Portuguese employees (!) were confronted with a wage freeze. Inflation would therefore be very useful to facilitate wage adjustment. He then showed how externally imposed wage growth – by applying 30,000 minimum wages agreed through collective bargaining – kept employment growth down in the period 1986-2009. In his discussion, Joop Hartog (University of Amsterdam) pointed out that the paper could make a better distinction between externally imposed wage growth and negotiated wage growth. After all, he cannot identify which companies were present at the negotiating table and which companies were bound because collective labour agreements were extended. Hartog furthermore notes that the criteria for administrative extension of collective labour agreements (e.g. majority condition, assessment of costs and benefits) seem quite reasonable, although of course it is key who is actually making this assessment.

DNB-IMF Conference

ally, Ernesto Villanueva (Banco de España) presented a paper about contract staggering in Spain. He shows that employees covered by collective labour agreements concluded before the collapse of Lehman Brothers in September 2008 ran a higher risk of losing their jobs than employees under 'updated' contracts, especially if these old contracts are of long duration and wages are closer to the minimum set in the collective labour agreements. In her discussion, Jante Parlevliet (DNB) asked questions about some important building blocks of the paper (in particular the exogeneity of the Lehman Brothers shock – after all, employment declined quite some time before that). Furthermore, she informed about the relative importance of contract staggering for the Spanish employment disaster; for the slow wage adjustment after the Lehman Brothers shock is also often mentioned as a large problem, just as the widespread use of temporary contracts and the impact of credit restrictions. As long-term contracts are customary in many countries‎, the general discussion focused on ways to make collective contracts responsive to economic conditions, including during the course of the contract.
With three countries' experiences, the third and last session was the most policy-oriented. 
Helge Berger (IMF) first presented IMF work, in particular recent Article IV reports, about the adjustment process in Spain. According to a sectoral wage regression, the recent labour market reforms would have a visible effect on wage adjustments, although discussant Marcel Jansen (Universidad Autonoma de Madrid) thought it would take some more time to definitely identify their full impact. Furthermore, the presentation gave a broad overview of the various challenges to the Spanish economy: high structural unemployment, a large influence of house prices on the economy, private sector debt overhang and, in particular, a chronic productivity problem.

DNB-IMF Conference

Gilbert Cette (Banque de France) discussed reform challenges of France, based on the work with, among others, Philippe Aghion. He discussed a variety of issues. That the French labour market is rigid - with high protection from dismissal, relatively high minimum wages, a large wedge between net and gross wages - is commonly known. In addition, Cette referred to rigid product markets, the deteriorating educational performance, low social mobility and high inequality of market incomes. Meanwhile, the current account balance has been going down slowly but steadily since 2005. Economists and international organisations have recommended a consistent range of reforms to France, but actual reforms are scarce. An exception according to Cette is a far-reaching liberalisation of the bus transportation market. Discussant Eric Bartelsman (VU University) observed that liberalisation of regular bus services is still relatively easy compared to adapting to the new technological challenge of self-driving buses. Considering the slow pace of reforms, Cette also commented shortly on the political aspects of the economy. In addition to the usual difficulties of reform – such as the occurrence of concentrated losses, but dispersed and uncertain revenues – in France the persistent misconception is that what is good for companies must be bad for employees.

Daniel van Vuuren (CPB) gave the concluding speech about the role the Netherlands Bureau for Economic Policy Analysis (CPB) plays in assessing structural reforms (including of election platforms). Van Vuuren started by outlining the origins and institutional setting of the CPB. Despite being funded by the Ministry of Economic Affairs its independence is widely acknowledged. Still, other countries are well advised to guarantee the financial independence of such a new institution.  Next, Van Vuuren described the CPB's role, from setting the agenda to providing neutral information, illustrated by four cases: young disabled persons and the Work-incapacitated Persons (Participation) Act, budgetary labour market policy based on the MicSim model, tax treatment of self-employed individuals and the increase of retirement age. Massimo Giuliodori (University of Amsterdam) in his discussion confirmed CPB's independent position by looking at estimation errors in budgetary forecasts (no bias). Furthermore, he warned that too large an expansion of the CPB's range of tasks could undermine its reputation as it is key that CPB has enough expertise to underpin its reform assessments.